Hello all,
we are facing some issues in optical property calculation of 2D heterostructures.
For monolayers we are able to analyse the optical properties, real and imaginary part of dielectric function etc accurately. But when it comes to heterostructure with a 25A' vacuum, we are facing issues.
usually for monolayers it will be comming like
writing wavefunctions
optical routines
imaginary and real dielectric function
recalculating local potential from charge density
direction 1
direction 2
direction 3.
but for heterostructure, we are getting following in the slurm.out file.
writing wavefunctions
optical routines
imaginary and real dielectric function
recalculating local potential from charge density
direction 1
Because of this, we are unable to extract the real and imaginary parts from the vasprun.xml file.
Could you please help us to resolve this issue?
We tried calculations by increasing the KPOINTS. But still we are getting the same results.
We are using the same INCAR for both monolayer and heterostructure optical porperty calculations.
Is there anything wrong with this approach. The iNCAR is as follows
#ISTART= 1
#ICHARG= 11
LOPTICS= TRUE
NBANDS= 200
ALGO = Fast
EDIFF = 0.000001
ENCUT = 300
IBRION = -1
#ISIF = 4
ISMEAR = -5
SIGMA= 0.01
#LORBIT = 11
LREAL = Auto
NELM = 200
NSW = 0
PREC = Normal
EDIFFG=0.01
LVDW = TRUE
IVDW = 12
Hoping for your suggestions,
Thanks in advance.
SCANMAT
Optical property calculation of 2D heterostructures
Moderators: Global Moderator, Moderator
-
- Newbie
- Posts: 28
- Joined: Mon Feb 01, 2021 1:53 pm
-
- Administrator
- Posts: 282
- Joined: Mon Sep 24, 2018 9:39 am
Re: Optical property calculation of 2D heterostructures
Hello,
please provide a minimal error report including input and output files.
If possible, can you provide a minimal system where the issue occurs?
please provide a minimal error report including input and output files.
If possible, can you provide a minimal system where the issue occurs?
-
- Newbie
- Posts: 28
- Joined: Mon Feb 01, 2021 1:53 pm
Re: Optical property calculation of 2D heterostructures
Hello,
Here I am attaching the POSCAR under study, OUTCAR and slurm file obtained after the optical property calculations.
INCAR file was attached in the previous mail.
Thanks and regards.
Here I am attaching the POSCAR under study, OUTCAR and slurm file obtained after the optical property calculations.
INCAR file was attached in the previous mail.
Thanks and regards.
You do not have the required permissions to view the files attached to this post.
-
- Administrator
- Posts: 282
- Joined: Mon Sep 24, 2018 9:39 am
Re: Optical property calculation of 2D heterostructures
Following line in slurm-47109.err indicates that you are out of memory on the device.
This means you cannot allocate 6 GB of memory on the GPU.
It might be that the first line indicates that increasing the allowed memory limit with "ulimit" failed.
I suggest to identify the origin of this ulimit-message and try to fix it in your batch script or system settings.
Alternatively, you could try a smaller unit cell by reducing the vacuum in c-direction.
Code: Select all
/var/spool/slurm/d/job47109/slurm_script: line 19: ulimit: max memory size: cannot modify limit: Operation not permitted
...
...
...
Out of memory allocating 6544179200 bytes of device memory
...
It might be that the first line indicates that increasing the allowed memory limit with "ulimit" failed.
I suggest to identify the origin of this ulimit-message and try to fix it in your batch script or system settings.
Alternatively, you could try a smaller unit cell by reducing the vacuum in c-direction.
-
- Newbie
- Posts: 28
- Joined: Mon Feb 01, 2021 1:53 pm
Re: Optical property calculation of 2D heterostructures
Thank you for the reply.
We tried using a reduced KPOINTS, and it worked.
Since this is a heterostructure, I couldn't further reduce the vacuum.
However, by reducing the KPOINTS, we resolved the memory problem.
Thank you very much for the help.
We tried using a reduced KPOINTS, and it worked.
Since this is a heterostructure, I couldn't further reduce the vacuum.
However, by reducing the KPOINTS, we resolved the memory problem.
Thank you very much for the help.